Skip to content

[SPARK-57562][SQL] Add benchmarks for the TIME data type #56809

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
shrirangmhalgi wants to merge 8 commits into
base: master
Choose a base branch
from
+603 −242
Open

[SPARK-57562][SQL] Add benchmarks for the TIME data type #56809

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
06591f3
[SPARK-57562][SQL] Add benchmarks for the TIME data type
shrirangmhalgi Jun 25, 2026
b5a2054
Add TIME cases to ExtractBenchmark, JsonBenchmark, and CSVBenchmark
shrirangmhalgi Jun 26, 2026
098bfd2
Address review: fix time generator formula, revert Mac results, updat…
shrirangmhalgi Jun 27, 2026
da3ac84
Benchmark results for org.apache.spark.sql.execution.benchmark.TimeBe…
shrirangmhalgi Jun 27, 2026
f0100c4
Benchmark results for org.apache.spark.sql.execution.benchmark.Extrac…
shrirangmhalgi Jun 27, 2026
15a0289
Benchmark results for org.apache.spark.sql.execution.datasources.json…
shrirangmhalgi Jun 27, 2026
96896fd
Benchmark results for org.apache.spark.sql.execution.datasources.csv.…
shrirangmhalgi Jun 27, 2026
be71268
Benchmark results for org.apache.spark.sql.HashBenchmark (JDK 17, Sca…
shrirangmhalgi Jun 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
50 changes: 25 additions & 25 deletions sql/catalyst/benchmarks/HashBenchmark-results.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,69 +2,69 @@
single ints
================================================================================================

OpenJDK 64-Bit Server VM 17.0.18+8-LTS on Linux 6.14.0-1017-azure
OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 9V74 80-Core Processor
Hash For single ints: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
interpreted version 2035 2040 7 263.8 3.8 1.0X
codegen version 3949 3952 3 135.9 7.4 0.5X
codegen version 64-bit 3419 3423 6 157.0 6.4 0.6X
codegen HiveHash version 2647 2663 23 202.8 4.9 0.8X
interpreted version 2044 2045 2 262.7 3.8 1.0X
codegen version 3953 3956 4 135.8 7.4 0.5X
codegen version 64-bit 3439 3443 5 156.1 6.4 0.6X
codegen HiveHash version 2749 2750 2 195.3 5.1 0.7X


================================================================================================
single longs
================================================================================================

OpenJDK 64-Bit Server VM 17.0.18+8-LTS on Linux 6.14.0-1017-azure
OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 9V74 80-Core Processor
Hash For single longs: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
interpreted version 2486 2492 8 215.9 4.6 1.0X
codegen version 5601 5602 1 95.8 10.4 0.4X
codegen version 64-bit 4175 4183 11 128.6 7.8 0.6X
codegen HiveHash version 3263 3264 0 164.5 6.1 0.8X
interpreted version 2492 2493 2 215.5 4.6 1.0X
codegen version 5630 5631 0 95.4 10.5 0.4X
codegen version 64-bit 4077 4078 1 131.7 7.6 0.6X
codegen HiveHash version 3146 3147 2 170.7 5.9 0.8X


================================================================================================
normal
================================================================================================

OpenJDK 64-Bit Server VM 17.0.18+8-LTS on Linux 6.14.0-1017-azure
OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 9V74 80-Core Processor
Hash For normal: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
interpreted version 2734 2748 20 0.8 1303.7 1.0X
codegen version 2511 2512 1 0.8 1197.4 1.1X
codegen version 64-bit 759 760 1 2.8 362.1 3.6X
codegen HiveHash version 4104 4105 1 0.5 1957.1 0.7X
interpreted version 2756 2758 2 0.8 1314.4 1.0X
codegen version 2574 2574 0 0.8 1227.2 1.1X
codegen version 64-bit 766 774 12 2.7 365.5 3.6X
codegen HiveHash version 4182 4185 5 0.5 1994.1 0.7X


================================================================================================
array
================================================================================================

OpenJDK 64-Bit Server VM 17.0.18+8-LTS on Linux 6.14.0-1017-azure
OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 9V74 80-Core Processor
Hash For array: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
interpreted version 1158 1158 0 0.1 8836.3 1.0X
codegen version 3648 3654 9 0.0 27829.0 0.3X
codegen version 64-bit 2699 2700 1 0.0 20595.1 0.4X
codegen HiveHash version 889 889 1 0.1 6779.7 1.3X
interpreted version 1389 1391 3 0.1 10595.8 1.0X
codegen version 3945 3945 0 0.0 30098.4 0.4X
codegen version 64-bit 2977 2977 0 0.0 22710.3 0.5X
codegen HiveHash version 1086 1088 2 0.1 8289.0 1.3X


================================================================================================
map
================================================================================================

OpenJDK 64-Bit Server VM 17.0.18+8-LTS on Linux 6.14.0-1017-azure
OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 9V74 80-Core Processor
Hash For map: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
interpreted version 0 0 0 101.5 9.9 1.0X
codegen version 247 248 0 0.0 60408.3 0.0X
codegen version 64-bit 193 193 0 0.0 47021.2 0.0X
codegen HiveHash version 30 30 0 0.1 7362.8 0.0X
interpreted version 0 0 0 102.5 9.8 1.0X
codegen version 201 201 1 0.0 49039.1 0.0X
codegen version 64-bit 157 157 0 0.0 38253.6 0.0X
codegen HiveHash version 16 17 1 0.3 3958.7 0.0X


1 change: 1 addition & 0 deletions sql/catalyst/src/test/scala/org/apache/spark/sql/HashBenchmark.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -125,6 +125,7 @@ object HashBenchmark extends BenchmarkBase {
.add("binary", BinaryType)
.add("date", DateType)
.add("timestamp", TimestampType)
.add("time", TimeType())

@MaxGekk MaxGekk Jun 27, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This adds a time column to the normal hash schema, but HashBenchmark-results.txt was reverted to upstream/master in this commit — it still has the old hardware header (17.0.18 / Linux 6.14) and no time column. Round 2 (727e46b) had this file correctly regenerated on CI, so the benchmark runs fine; this commit just dropped the regenerated results while keeping the schema change.

Your note mentions regenerating "all four" results files (the sql/core ones); HashBenchmark lives in sql/catalyst and is a fifth modified benchmark whose results were missed. As committed, the benchmark code and its results file are out of sync. Please regenerate HashBenchmark-results.txt on the CI runner too (so it reflects the new time column), or drop this HashBenchmark schema change if it's out of scope for the PR.

test("normal", normal, 1 << 10, 1 << 11)

val arrayOfInt = ArrayType(IntegerType)
Expand Down
Loading